شناسايي تظاهرات احساسي درد در تصاوير چهره

Recognition of Pain Expression in Facial Images

A facial expression discrimination system has been developed which uses grayscale images and classifies the pains expression against the other facial expressions including: Neutral, Happiness, Anger and Surprise expressions. Principal Component Analysis (PCA), Zernike Moments and Gabor Wavelet Transform have been used for feature extraction and Support Vector Machine (SVM), Linear Discriminant Analysis (LDA) and Feed Forward Neural Network have been used for feature classification. A database including 476 face images from 30 persons has been used in this research. The result of classification is 77.44% for severe pain, 73.71% for moderate pain and 71.15% for low pain expression.